DESCRIPTION (provided by): Generally speaking, substance use tends to peak in emerging adulthood and then diminish after that, becoming more adult-like by early adulthood (the late 20s) a process referred to as maturing out. As a result, these use habits at age 30 are very good predictors of future use and abuse for many years to come; and yet, they have received very little empirical attention to date. These use trajectories also differ by race. Black adolescents use less than white adolescents, but the difference diminish by the mid-20s, and reverses in the late 20s, especially for marijuana and tobacco use. In addition, reports of abuse, among users, are higher for blacks than whites (referred to as a racial crossover effect). This difference contributes to health disparities in the US-e.g., higher rates of certain cancers in blacks. A similar pattern exists in sexual risk behavior, as young black adults are less likely to be involved in a steady relationship and more likely to have multiple partners- one reason why HIV infection rates are high in blacks. A general assumption is that these racial differences in risky behavior reflect increased stress among blacks, including the stress associated with perceived racial discrimination. There is also relatively little support for this hypothesis, howevr, because studies of racial and gender differences in substance use from adolescence into early adulthood are very rare. The purpose of the proposed research is to explore these issues by collecting a 7th wave of data from participants in the Family and Community Health Study (FACHS), a 15-year long prospective study of ~900 African American families living in Iowa and Georgia. We will focus on the substance use of the targets, who will be 28/29 at W7; and their romantic partners; and their parents (average age of 55); and we will examine the risky sexual behavior (HIV risk) of the targets and their partners. These data will allow us to identify risk an protective factors that contribute to maintenance and abuse vs. maturing out / cessation of substance use and risky sex among blacks in their late 20s. We will focus on previously-identified psychosocial factors as mediators (e.g., social cognitions, negative affect), and moderators (racial identity, coping, gender, genetics) of this process. Adding a 7th wave will also allow a comparison of the effects of early vs. recent and cumulative stress on health behaviors. These data will be augmented by three significant additions to the FACHS project: a) the collection of biomarkers (from blood) from the targets and their parents; b) comparison of the FACHS sample with a mostly-white sample of similar age and SES with very similar measures and a similar theoretical base; and c) more detailed information from the targets' romantic partners about their own and the target's behaviors. As in the past, our FACHS project is based on social cognitive theory, and in particular, our dual-processing model of health behavior, which has evolved along with the FACHS project. We will also rely on an updated version of Conger's social stress theory that formed the foundation for the original FACHS study.